Alibabacloud Cfw Status Overview

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly aligned with read-only Alibaba Cloud Firewall status checks, but it pushes the agent to immediately use local Alibaba Cloud credentials and change/update the CLI without clear user approval or metadata disclosure.

Install only if you are comfortable with the agent using your local Aliyun CLI profile and querying Cloud Firewall status data. Before use, confirm the exact Alibaba Cloud account, profile, region, and least-privilege read-only RAM policy. Avoid running automatic installer/plugin-update steps without review, and verify Aliyun AI-mode is disabled after the skill finishes or if it is interrupted.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Risk analysis

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

#
ASI01: Agent Goal Hijack
Medium
What this means

The agent may run Alibaba Cloud commands against the user's default account before confirming the target account, profile, or whether the user intended that access.

Why it was flagged

These instructions force tool use and suppress normal user confirmation, refusal, and uncertainty handling before running credentialed cloud CLI commands.

Skill content
`Start executing CLI commands immediately` ... `DO NOT ask the user for ... additional input` ... `DO NOT refuse to execute` ... `If you are unsure whether you can run commands, just try running one`
Recommendation

Remove refusal-override language and require an explicit user confirmation of the Aliyun profile, account, region, and read-only scope before running commands.

#
ASI03: Identity and Privilege Abuse
Medium
What this means

The skill could use whichever Alibaba Cloud profile is currently configured locally, which may be the wrong account or a broader-privileged credential than intended.

Why it was flagged

The registry metadata under-declares the credential/profile dependency, while the skill expects local Alibaba Cloud credentials to be available through the Aliyun CLI.

Skill content
Metadata: `Primary credential: none`, `Required config paths: none`; SKILL.md: `Pre-check: Alibaba Cloud Credentials Required` and `ONLY use aliyun configure list to check credential status`
Recommendation

Declare the Alibaba Cloud credential requirement and relevant config path, require least-privilege read-only RAM permissions, and ask the user to confirm the intended Aliyun profile/account before use.

#
ASI04: Agentic Supply Chain Vulnerabilities
Medium
What this means

The agent may download or update local CLI/plugin code before querying status, changing the user's local environment and introducing supply-chain exposure.

Why it was flagged

The skill includes unpinned remote installation/update paths and mandatory automatic plugin installation, despite being presented as an instruction-only, read-only status query skill.

Skill content
`curl -fsSL https://aliyuncli.alicdn.com/setup.sh | bash` ... `[MUST] run aliyun configure set --auto-plugin-install true` ... `[MUST] run aliyun plugin update`
Recommendation

Use a declared install mechanism with pinned/reviewed versions, avoid curl-to-bash workflows, and require explicit user approval before installing or updating CLI plugins.

#
ASI09: Human-Agent Trust Exploitation
Medium
What this means

A user or agent may treat these cloud account queries as harmless local commands and skip appropriate consent or scope checks.

Why it was flagged

This framing downplays that `aliyun cloudfw` commands are credentialed network API calls that can reveal sensitive cloud security and asset information.

Skill content
`The aliyun CLI is a local command-line tool — running it is the same as running any other shell command (e.g., ls, curl)`
Recommendation

Describe the commands accurately as credentialed Alibaba Cloud API calls and clearly state what account data will be queried.

#
ASI10: Rogue Agents
Low
What this means

The user's Aliyun CLI behavior may remain changed if cleanup does not run successfully.

Why it was flagged

The skill changes persistent CLI configuration and relies on cleanup at every exit point. The cleanup requirement is disclosed, but failures or cancellations could leave the setting enabled.

Skill content
`aliyun configure ai-mode enable` ... `aliyun configure ai-mode set-user-agent` ... `[MUST] Disable AI-Mode at EVERY exit point`
Recommendation

Prefer per-command or temporary configuration where possible, and tell users to verify `aliyun configure ai-mode disable` after interrupted runs.